The Physics of Dark Matter Detection: Challenges and Emerging Techniques

Abstract

Dark matter remains one of the most compelling problems in contemporary physics, inferred gravitationally across galactic to cosmological scales yet undetected through non-gravitational interactions. This paper surveys the physics foundations of dark matter detection, synthesizing theoretical motivations with the experimental landscape and the statistical methodologies that connect models to data. We first outline the principal candidate spaces—from GeV–TeV weakly interacting massive particles (WIMPs) to sub-GeV light dark matter, axions and axion-like particles, sterile neutrinos, and more exotic composite or ultralight fields—and describe the corresponding interaction channels with baryonic matter and photons. We then review direct-detection strategies that seek nuclear or electronic recoils using cryogenic calorimeters, dual-phase noble time projection chambers, and low-threshold semiconductor and superconducting sensors that leverage phonon, magnon, plasmon, and polar-material responses. Particular emphasis is placed on backgrounds (radiogenic, cosmogenic, and solar/atmospheric neutrinos), material radiopurity, shielding and self-shielding, fiducialization, and discrimination observables, culminating in the “neutrino floor” and techniques proposed to surpass it, such as directional detection, target complementarity, and temporal signatures. Complementary approaches—indirect detection via high-energy photons, cosmic rays, and neutrinos; collider searches that exploit missing-momentum topologies; and precision astrophysical and cosmological probes including microlensing, pulsar timing, stellar streams, and cosmic microwave background anisotropies—are integrated into a unified framework of cross-validation and parameter-space coverage. Across these fronts, we highlight emerging techniques: quantum-enhanced sensing, dielectric and resonant haloscopes for axions, LC-circuit readout, single-electron and single-optical-phonon sensitivity, underground accelerators for calibration, and machine-learning–assisted background modeling and likelihood-free inference. We conclude with a forward-looking assessment of experimental roadmaps, synergy across detection modalities, and the statistical and metrological advances required to convert incremental sensitivity gains into robust discovery potential.